Research has shown that despite significant rigor and intensity, some children with an autism spectrum disorder fail to make significant gains in response to behavioral treatment (e.g., Sherer &Schreibman, 2005). The behavioral characteristics of children who excel versus those that do not are not well understood. The biological characteristics of such treatment responders and nonresponders are completely unknown. In order to determine what behavioral and biological factors predict treatment responsiveness, consistency regarding various aspects of treatment must be provided. For all children the type of treatment, age at which treatment is administered, and length of treatment must be consistent. Therefore, a Treatment Core (TxC) has been established to implement an evidence-based treatment for toddlers with autism participating in the ACE projects. A specific, manualized treatment, the STAR Program (see Appendix A), has been chosen as the foundation for the treatment. This curriculum incorporates evidence-based behavioral methods, including discrete trial teaching, pivotal response training, and teaching within functional routines, that are of documented effectiveness. This curriculum will be adapted to accommodate 2-year-old children and to include additional social goals and developmental strategies. Recently, a developmental, social-pragmatic intervention that incorporates both developmental and naturalistic behavioral strategies has been examined as a parent education adjunct to the STAR curriculum. This program incorporates early developmental strategies such as Responsive Teaching and Floor Time/DIR, which encourage joint attention, social responsiveness, and engagement in children with autism. We believe, and preliminary data agree, that this combination of curricula and strategies will provide an effective and consistent treatment for children with autism. Overall, the Treatment Core (TxC), has two main goals: The first is to provide state-of-the art behavioral treatment to all participants who meet provisional criteria for autism at 2 years. In order to meet this goal, the TxC will ensure consistent application of treatment practices by training all study personnel in treatment protocol to mastery level. Additionally, the TxC will ensure fidelity of implementation of the treatment protocol throughout the funding period. The second goal is to provide a quantitative index of level of response to treatment for each at-risk toddler for use in predictive analyses in Projects 1-4. As such, the TxC will collect data throughout treatment to assess learning rate and speed of progress in each curriculum area. Finally, the TxC will establish an overall response-to-treatment profile for each study toddler in the domains of symptom severity, cognitive, language, and social behavior. These profiles will be used by each project as well as the Integrated Biostatistics and Bioinformatic Analysis Core to determine the profiles of treatment responders and nonresponders.